You are on page 1of 21

Interpretation of Subgrade Reaction from

Lateral Load Tests on Spun Piles in Soft Ground

Ir. Tan Yean Chin, Ir. Dr. Gue See Sew & Ir. Fong Chew Chung
G&P Geotechnics Sdn Bhd, Kuala Lumpur, Malaysia


This paper presents the results and interpretation of a lateral load test
on a fully instrumented spun pile in soft ground for the land viaduct
section of a high speed train project in Alor Pongsu area. The lateral
load test of the spun piles was carried out on a pile with fixed pile
penetration to measure the lateral response of single pile and was tested
by jacking against a reaction pile in accordance to ASTM D3966 – 1995.

Palmer and Thompson’s (1948) formula for subgrade reaction (kh) was
used to interpret the instrumented results of the lateral load test.
Comparison with Davisson’s proposed kh with undrained shear strength
(su) was also made. The results show a correlation with the coefficient of
subgrade reaction (nh) which removes the effect of subsoil depth and
pile diameter. At depths up to 4m, Davisson’s prediction is conservative
but actual nh is much higher due to overconsolidation of subsoil at
shallow depths. For deeper subsoil, it slightly overpredicts due to
passive resistance of the subsoil not yet fully mobilised. The back
analysed Davisson’s constant, is found to be 50 and is proposed for
local soft ground conditions.


The site is part of a project for a high speed railway traversing the northern region of
Malaysia. Figure 1 shows the location of the site. A portion of the railway of about
28km length is supported by land viaducts due to it’s cost effectiveness and shorter
construction duration as compared to piled embankment. As part of the pile
verification tests, a fully instrumented preliminary test pile was installed and tested to
validate the lateral pile performance.

This paper presents the interpretation of the subgrade modulus from the results of
the fully instrumented lateral pile test. Subgrade reaction approach was used as the
design of the land viaduct is based on pile bent approach where the foundation
design is based on integral bridge design concept without bearings and minimum
joints. This is a variation on the column bent approach where the supporting columns
and foundation are replaced with individual supporting piles (Tonias, 1991). Hence,
the bridge superstructure designer uses the subgrade reaction values as design input
to analyse and design the forces in the land viaduct superstructure.

Site Location

Figure 1: Location of the Site


2.1 General

There are many methods of analysing the response of a laterally loaded pile. These
methods can be categorised into subgrade reaction approach and elastic continuum
approach. Subgrade reaction approach has been initially proposed by Palmer &
Thompson (1948) and subsequently further developed by Reese and Matlock (1956).
Further advancements lead to the development of p-y curves and are commonly
used to model the non-linear pile and soil behaviour. These have been described by
McClelland and Focht (1958) and Davisson and Gill (1963). Further details and
descriptions of p-y curves are summarised by Reese & Van Impe (2001).

In this paper, the subgrade reaction approach is used to analyse the results of the
instrumented test pile. This method is commonly used by bridge engineers to model
and analyse pile bent structures. The historical development of subgrade reaction
method begins with Winkler in 1867 modelling a beam on soil and subsequently
adopting it to model embedded piles by others. It characterises the soil as a series of
unconnected linearly-elastic springs in response to loading on the pile. In the model,
the horizontal pressure (p) and the corresponding deflection at a point (y) is related
by a horizontal modulus of subgrade reaction (kh):

p = kh y
p = soil reaction per unit length of pile
y = pile deflection
kh = subgrade reaction in units of force/length2

Palmer & Thompson (1948) subsequently expressed the above equation in the form

kh = k L ⎜ ⎟
kL = value of kh at pile toe (z = L)
n = a coefficient greater than zero

For sands and normally consolidated clays under long term loading, n is taken as
unity. For overconsolidated clays, n is commonly taken as zero. However, the
commonly used form when n = 1 and adopting a variation of kh with depth:

k h = nh ⎜ ⎟
⎝d ⎠

Where nh is the coefficient of subgrade reaction and d is pile diameter.

This applies to cohesionless soils and normally consolidated clays where these soils
indicate increasing strength with depth due to increase in overburden pressure.

(1) (2)

Figure 2 : Subgrade Reaction Model of (1) Actual Soil Reaction on Pile & (2) Elastic
Spring Model of Soil Reaction - after Prakash& Sharma (1990)

2.2 Cohesive Soils

A number of empirical correlations for kh in cohesive soils have been proposed by

Broms (1946), Skempton (1951) and Baguelin (1978).

Broms (1946): k h = 1.67 * E50 / d

Skempton (1951): k h = (80 − 320) * Cu / d

E50 = secant modulus at half the ultimate stress in undrained test
Cu = undrained shear strength

d = pile diameter

However, for preliminary design before any verification test, a conservative approach
suggested by Davisson’s (1970) was used:

k h = 67

For cohesive soils with kh increasing linearly with depth, kh is usually expressed in
the form of k h = nh * z / d . Table 1 summarises the typical values of nh for cohesive
soils by various authors.

Soil Type nh (kN/m3) Reference

Soft NC Clay 163 – 3447 Reese & Matlock (1956)

271 – 543 Davisson & Prakash (1963)
NC Organic Clay 179 - 271 Peck & Davisson (1962)
179 – 814 Davisson (1970)
Peat 54 Davisson (1970)
27 – 109 Wilson & Hilts (1967)
Loess 7872 – 10858 Bowles (1968)

Table 1: Typical Values of nh for Cohesive Soils

2.3 Cohesionless Soils

For piles in cohesionless soils, Terzaghi (1955) proposed :

nh = A (tons/ft3)

Typical values of dimensionless factor A is shown in Table 2

Relative Density Loose Medium Dense

Range of values of A 100 – 300 300 – 1000 1000 – 2000
nh , Dry moist sand (kN/m3) 2425 7275 19400
[ton/ft3] [7] [21] [56]

nh , Submerged sand (kN/m3) 1386 4850 11779

[ton/ft3] [4] [14] [34]

Table 2: Typical Values of nh for Cohesionless Soils - after Terzaghi (1955)


The land viaduct consists of multiple spans, each span is typically 15m long between
piers. Each 10.5m wide pier is supported by six 600mm diameter high strength
circular spun piles. The spacing between the piles is three times the pile diameter. In
the pile bent design, the spun pile head is directly cast into the crosshead. Figure 3
shows a typical cross section of the land viaduct. This pile bent design is limited to
3m high between the ground level and the soffit of the crosshead. Figure 4 shows
part of the completed land viaduct.

Figure 3: Typical Cross Section

Figure 4: Completed Land Viaduct


4.1 General Geology

Most of the route of the project traverses near the coastal areas of western
Peninsular Malaysia. These areas are often underlain by marine alluvium formation.
Figure 5 shows the existing railway on the geological map.

The railway alignment traverses through different materials ranging from soft marine
clay to dense residual soil. In the test pile area, the subsoil is generally of soft marine
clay from Quaternary and Holocene periods. Table 3 describes the geology of these
two sections. The test pile is located in Alor Pongsu, near Kamunting, Perak and is
also indicated in Figure 5.

Section & Location Formation Age Lithology

Gravel, Sand, Clay
to Superficial Deposit Quaternary
Parit Buntar

Table 3 : General Geology of the Site

Marine and continental deposits

Shale, Sandstone, conglomerate

Cross-bedded Sandstone with


Interbedded Sandstone Siltstone &


Phyllite, Slate & Shale with

subordinate sandstone & schist

Phyllite, Slate, Shale & Sandstone

Phyllite, Schist & Slate with locally

Schist, Phyllite, Slate & Limestone
Sandstone/metasandstone with
subordinate siltstone, shale.

Location of Test Pile

Figure 5: General Geology of the Site

4.2 Geotechnical Characterisation

Several series of subsurface investigation were carried out around Site A. These
include boreholes, piezocones, field vane shear, mackintosh probes and associated
laboratory tests. Figure 6 shows a few of the boreholes near the preliminary test pile
while Figure 7 shows the laboratory test results on the soil samples collected.

Location of Test Pile

Figure 6: Borehole Logs

From the borelogs, the test pile site is characterised as highly compressible marine
alluvium with soft clay thickness of 15m to 30m. Soil classification tests show that the
subsoil is generally of high to extremely high plasticity clay. In addition, it has high
compressibility as shown by the compression ratio of 0.20 to 0.35 and re-
compression ratio of 0.055 (Figure 8). Stress history of the subsoil show that the
upper 3m is overconsolidated while the clay stratum is generally slightly
overconsolidated (Figure 9) and very soft as shown by the undrained shear strength
profile (Figure 10). In such difficult subsoil conditions and where high fills is needed if
embankment is used, land viaduct was adopted as a more cost and time effective

method of construction as compared to general ground treatment methods such as
piled embankment.
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

100 100

High Plasticity
Low Plasticity MBH-B10

Very High
Extremely High

90 90 MBH-B11
80 80
70 70 P9-MABH-5


CV A Line 60 P9-MAHA-3

PI = 0.73(LL-20) P9-MAHA-5
Plasticity Index,

50 CH 50

40 40 P9-MAHA-14
30 30 P9-MAHA-22
20 20
10 10
0 0

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Liquid Limit,LL
Liquid Limit, LL(%)

Figure 7: Soil Characterisation Test Results

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

0 0 0 0

5 5 5 5

10 10 10 10
Depth (m)
Depth (m)

15 15 15 15

20 20 20 20

25 CR=0.20 25 25 25

30 30 30 30

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
Compression Ratio, CR = cc / (1 + eo) Recompression Ratio, RR = cr / (1+ec)

Figure 8: Compressibility of Subsoil

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0 0

0 0
OCR = 3.5
1 1
OCR = 2.5
2 2
OCR = 1.5 Legend
3 3
5 MBH-B10
5 4 4
5 5
6 6
OCR = 1.2 MBH-B13
7 7
10 8 8
10 9 9 P9-MABH-4
10 10 P9-MABH-5
Depth (m)

11 11 P9-MABH-6
Depth (m)

12 12 P9-MAPZ-4
13 13 P9-MAPZ-5
15 14 OCR = 1.4 14 P9-MAPZ-6
15 15 MPZ-B8
16 16
20 17 17
18 18
Cvo = 1.5 m2/yr 20 19 19
20 20
21 21
25 22 22
23 OCR = 1.8 23
24 24
25 25
26 26
27 27
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Cvo, m2/year Over Consolidation Ratio

Figure 9: Soil Stress History and Consolidation Parameters

0 10 20 30 40 50 60 70 80 90 100

0 0

1 1

2 2

3 3

4 4

5 5
Depth (m)

6 6

7 7

8 8

9 9

10 10

11 Mackintosh Probes 11
Field Vane
12 Interpreted Su Line 12

13 13

0 10 20 30 40 50 60 70 80 90 100
Su (kPa) / Mackintosh Probe Blow Count per foot
Figure 10: Undrained Shear Strength Profile


5.1 Test Setup

The lateral pile test was carried out according to ASTM D3966 – 1995 similar to the
test setup for testing two piles simultaneously. The test pile is jacked against a similar
or much stiffer reaction pile at some distance away. For this test setup, the distance
between the two piles is 1.8m centre-to-centre (three pile diameters). Figure 11
shows the lateral pile test setup.

The preliminary test pile consists of a 600mm diameter grade 80 circular spun pile
with wall thickness of 100mm. The pile is reinforced with 14 numbers of 10.7mm
diameter PC strands with effective prestress of 7.0 MPa. Figure 12 shows the details
of the test pile. The pile was driven to 36m depth and not to refusal as the aim of the
test is to measure the lateral response of the pile which generally becomes
insignificant after about 10 times the pile diameter.

The reaction pile is stiffer due to the thicker wall thickness of 120mm. The reaction
pile was initially used for the preliminary axial compression test and the manufacturer
had overcast the wall thickness to ensure the minimum 100mm thick sound concrete
(excluding any laitance) was obtained after spinning. The stiffer spun pile was used
as reaction pile as it would provide a better reaction mass due to it’s higher stiffness.

Signal cables of
VWSGs, LVDTs & Reference Frame
Load Cell to

203 x 203 x 46 UC

50t VW Load Cell

100t Hyd. Jack


Test Pile Reaction Pile

Figure 11: Lateral Pile Test Setup

Figure 12: Details of Test Pile

5.2 Instrumentation Details

Instrumentation of the test pile consists of 12 levels of strain gauges at specified

levels and inclinometer in the hollow middle of the spun pile as shown in Figure 13.
The first six levels of strain gauges from ground surface have four strain gauges in a
cross-axis layout in anticipation that some of the strain gauges may be damaged
during pile installation. The remaining levels are instrumented with two strain gauges
per level as the driving stresses at these levels are expected to be much less
compared to those near the impact point. All strain gauges were welded to the PC
strands of the spun pile from the outer pile perimeter and housed in a protective
76x38x6.7kg/m C-Channel housing. Figure 14 shows the C-Channel housing on the
test pile prior to pile installation.

The inclinometer is a 75mm diameter inclinometer tube installed centrally in the

hollow centre of the spun pile with the aid of spacers. Subsequently, the void
between the inclinometer tube and inner diameter of spun pile was grouted with
bentonite –cement mix in the ratio of 1 : 7 to hold the inclinometer tube in place.
Sample cylinders of the mix were cast during mixing to check the in-situ strength at
14 days when the pile test was carried out. The results on three cylinders showed an
average unconfined compressive strength of about 633 kPa, which approximately
corresponds to undrained shear strength of about 316 kPa and is sufficient for the
purposes of the load test.

Test Pile Reaction Pile

Figure 13: Elevation View of Testing Setup and Instrumentation Details

Figure 14: C-Channel Protection of Strain Gauges on Spun Piles (Lee, 2008)

Test Pile

Figure 15: View of Setup during Testing

Figure 16: Top View of Test Pile with Inclinometer Tube


6.1 General

For this test, the pile was tested to a test load of 180 kN. Although the lateral pile
working load is 50kN, the pile was tested beyond several times it’s working load to
observe the failure behaviour. The following are the pile design criteria from the
bridge designer:

Lateral Pile Working Load (LPWL) 50 kN

Minimum Test Load (Lateral) 100 kN
Allowable Pile Head Deflection at 1 x LPWL 25 mm
Allowable Pile Head Deflection at 2 x LPWL 50 mm
Maximum Pile Service Moment 195.7 kNm
Table 4: Pile Design Criteria

The lateral load test was generally conducted in load increments of 10 kN to allow
the inclinometer to register deflection along the pile length. Figure 17 shows the load
schedule profile of the lateral pile test.



Legend :
Applied Load Duration








0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500
Duration (Minutes)

Figure 17: Load Schedule Profile

6.2 Results of Lateral Load Test

Figure 18 shows the results of the pile head deflection vs applied lateral load of the
test pile. The results generally show that at one lateral pile working load (50kN), the
pile head deflection was about 12.5mm and at twice LPWL (100kN), the pile head
deflection was about 31.3mm. Both results show that the pile design was within the
design deflection criteria.

As this was a sacrificial test pile, the lateral load was further increased to 120 kN for
the first cycle and to 180 kN for the second cycle to observe the lateral deflection
behaviour. At 180 kN, the loading portion of the load settlement curve still shows
approximate linear behaviour with no signs of yielding. In addition, the unloading
portion of the curve also shows linear rebound although the residual deflection is
quite large which is quite common for lateral pile in soft ground. This was probably
due to passive resistance of the soil still exerting on the pile body after the applied
load was removed.

Figures 19 and 20 show the deflection profile along the test pile body for the first and
second load cycle respectively. Generally, the test pile under applied lateral load
behaves like a free head pile which was characterised by the upper curved deflection
profile up to a depth where there is negligible deflection at about 8m depth. Although
the inclinometer still shows some deflection after 8m depth, this could be due to the
accuracy of the inclinometer which has accuracy of ±0.1mm.



Legend :
Pile Top Deflection







0 10 20 30 40 50 60 70 80 90 100

Figure 18: Pile Top Deflection vs Lateral Load

0 10 20 30 40 50

0 0
1 1
2 2
3 3
4 4
5 5
6 Legend 6
7 7
8 P=20kN 8
Depth (m)

9 P=30kN 9
10 P=40kN 10
11 11
12 P=70kN 12
13 P=80kN 13
14 P=90kN 14
15 15
16 P=120kN 16
17 17
18 18
19 19
20 20

0 10 20 30 40 50

Lateral Deflection (mm)

Figure 19: Deflection Results from Inclinometer – 1st Load Cycle

0 10 20 30 40 50 60 70 80

0 0
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 Legend 8
Depth (m)

9 P=0kN 9
10 10
11 P=90kN 11
12 P=120kN 12
13 P=130kN 13
14 14
15 P=160kN 15
16 P=170kN 16
17 P=180kN 17
18 18
19 19
20 20

0 10 20 30 40 50 60 70 80

Lateral Deflection (mm)

Figure 20: Deflection Results from Inclinometer – 2nd Load Cycle


From the test results, the subgrade reaction was interpreted assuming a fixity point
on the pile at 8m and a linearly varying soil pressure profile was determined for each
applied load. At each depth, the soil pressure was interpolated from the linear soil
pressure profile and the subgrade reaction (kh) was determined from the division of
soil pressure by the horizontal deflection at each depth respectively.

The results were plotted in Figure 21 alongside the initial calculated subgrade
reaction from Davisson’s (1970). Generally, the subgrade reaction profiles show a
peak at about five to six metres depth except for initial loading up to 20kN where the
profile is dissimilar due to the horizontal soil pressures not fully mobilised.

Figure 22 shows the plot of interpreted subgrade reaction with applied load for
several locations along the pile which indicates the subgrade reaction has mobilised
to a peak at applied loads of about 100kN to 120kN before softening in response.

0 0

1 P=10kN
2 2
3 P=80kN 3
Depth (m)

4 P=120kN 4
5 P=160kN
6 Davisson's 6

7 7

8 8

0 500 1000 1500 2000 2500 3000 3500

Interpreted Subgrade Reaction, kh (kN/m3)

Figure 21: Interpreted Subgrade Reaction

3000 3000

Graph 1
Interpreted kh, (kN/m3)

2000 2000

1000 1000

0 0

0 20 40 60 80 100 120 140 160 180 200

Applied Load (kN)

Figure 22: Interpreted Subgrade Reaction vs Applied Load

From the above plot, there is no visible trend of the subgrade reaction reaching peak
then strain softening as the load increases. However, the results were plotted in the
form of coefficient of subgrade reaction ( nh ) with depth for both load cycles and this
removes the effect of pile diameter and subsoil depth. These were presented in
Figure 23 with the initial predicted k h converted from Davisson’s (1970) plotted

In this plot, it can be seen that there is a general trend between the coefficient of
subgrade reaction with subsoil depth. In addition, most of the plots are within a
narrow range for most of the applied lateral load, suggesting the measured range of
nh values are closely within this range.

There is also a fair correlation between the predicted (Davisson’s -1970) and the
measured coefficient of subgrade reaction. At shallow depths, nh tends to be
underestimated significantly. However, this is probably due to overconsolidation of
the subsoil stratum near the ground surface as the nh profile is reminiscent of the
OCR profile.

In the first load cycle, the subgrade reaction starts to be fully mobilised after applied
lateral load of about 30kN. At 20kN, the profile indicates a partial mobilisation. At
depths up to about 4m, Davisson’s prediction is conservative but after 4m, it is
slightly overpredicted. Overprediction at shallow depths is due to conservativeness in
the selection of design line for the undrained shear strength. However, at deeper
depths, the slight overprediction could be due to passive resistance not fully
mobilised at that depth. The average nh value for this site was found to be about
200 kN/m3.

In the second cycle, the general profile is similar to the first cycle but the nh values
have reduced indicating a softer response of the soil to loading and correspond with
the expected behaviour in Figure 22. This could be due to the compressed soil after
the first cycle and have not rebound back to it’s original state after fully mobilising
passive resistance and possible yielding.
0 0

1 1

2 Legend 2
3 P=30kN 3
Depth (m)

4 P=70kN 4
5 5
6 P=140kN 6
7 P=180kN 7
Predicted nh

8 8

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200

Interpreted Coefficient of Subgrade Reaction, nh (kN/m3)

Figure 23: Interpreted Coefficient of Subgrade Reaction nh

Back analyses of the results were also carried out to determine Davisson’s constant
for the site. Figure 24 shows the results of the back analyses using the interpreted
k h . For this site, the proposed Davisson’s constant is 50 as compared to 67
proposed by Davisson. This constant is more conservative and is proposed for soft
ground in local ground conditions.

Overall, the initial prediction using Davisson’s method shows fairly good estimate of
the coefficient of subgrade reaction and subsequently the subgrade reaction values
after taking into consideration the pile size and depth of subgrade.

0 0

1 1

2 P=10kN 2
3 P=50kN 3
Depth (m)

4 4
5 P=130kN 5
6 P=170kN 6
Average Dc Value

7 7

8 8

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160

Interpreted Davisson's Constant with kh, DC

Figure 24: Back Analyses of Davisson’s Constant (Dc) with kh


A lateral load test was carried out on a fully instrumented preliminary (sacrificial) test
pile to verify the lateral performance of 600mm diameter spun pile for the land
viaduct section of a high speed train project. The land viaduct is built on soft marine
clay which is highly compressible and with low undrained shear strength. Each of the
pier in the land viaduct is supported by six numbers of 600mm diameter high strength
circular spun piles spaced at three times the pile diameter. The instrumented lateral
pile test was carried out according to ASTM D3966 – 1995 to verify the pile
performance under lateral loading. The results of the test are further analysed to
compare the initial prediction using subgrade reaction approach and recorded

The test pile was tested to maximum test load of 180kN in two cycles, well beyond
the lateral pile working load (LPWL) of 50kN and proof load of twice the LPWL of
100kN. At LPWL, the maximum deflection was 12.5mm and at twice LPWL, the pile
head deflection is about 31.3mm. The test results validate the design of the spun pile
under lateral loading and deflection were within specified limits.

The test results were further analysed to interpret the subgrade reaction profile ( k h )
along the pile depth. Generally, interpreted subgrade reaction profile shows a peak at
about five to six metres depth for all loading except for initial loading up to 20kN. This
is due to the horizontal soil pressures not fully developed. In addition, the initial
interpretation using subgrade reaction does not show any visible trend with applied
lateral load.

However, when the results are plotted in the form of coefficient of subgrade reaction
( nh ) with depth, it show a fair correlation with predicted nh using Davisson’s method.
At shallow depths, nh tends to be underestimated significantly due to
overconsolidation of subsoil stratum near the ground surface. At depths up to 4m,
Davisson’s prediction is conservative but slightly overpredicts after 4m depth due
passive resistance not fully mobilised at those depths. The average nh is 200 kN/m3
for this site.

Back analyses of Davisson’s constant for this project site show a proposed value of
50 compared to original value of 67 by Davisson. This is more conservative and is
proposed for soft ground in local ground conditions. Overall, the initial prediction
using Davisson’s method shows fairly good agreement. However, for soft ground in
Malaysia, the authors proposed a constant of 50 (ie. k h = 67*Su/d).


Ahmadi, M.M. & Ahmari, S. (2007), “A Simple Approach to Investigate Laterally Loaded Pile-
Soil Interaction,” Proceedings of 16th Southeast Asian Geotechnical Conference,
Subang Jaya, Malaysia.

Baguelin, F., Jezequel, J.F. & Shields, D.H. (1978), The Pressuremeter and Foundation
Engineering, Clausthal: Trans Tech Publications.

Broms, B. B. (1964). “Lateral Resistance of Piles in Cohesive Soils.” Journal of Soil

Mechanics and Foundations Div., Vol.90(2), 27–64.

Chen, L.T. & Poulos, H.G. (1996) “Pile Response due to Unsupported Excavation-Induced
Lateral Soil Movement”, Canadian Geotechnical Journal, Vol.33 No.4, p. 670-677.

Chen, L.T. & Poulos, H.G. (1997) “Piles Subjected to Lateral Soil Movements”, Journal of
Geotechnical and Geoenvironmental Engineering, Vol.123 No.9, pp.802-811.

Coutinho, R.Q., Horowitz, B., Soares, F.L. & Braga J.M. (2005), “Steel Pile under Lateral
Loading in a Very Soft Clay Deposit,” 16th International Conference on Soil
Mechanics and Geotechnical Engineering, Vol. 4, Osaka.

Davisson, M.T. (1970), “Lateral Load Capacity of Piles,” Highway Research Record,
Transportation Research Board, Washington, DC.

Kucukarsalan, S., Banerjee, P.K. & Bildik, N. (2003), Inelastic Analysis of Pile Soil Structure
Interaction, Engineering Structure: 1231-1239

Lee, S.K. (2008), Photos of Instrumentation of Spun Pile, Glostrext Technology Sdn Bhd,

Mosher, R.L. & Dawkins, W.P. (2000), Theoretical Manual for Pile Foundations, US Army
Corps of Engineers, Washington.

Palmer, L.A. & Thompson, J.B. (1948). The Earth Pressure and Deflection Along the
Embedded Lengths of Piles Subjected to Lateral Thrust. Proceedings of the Second
International Conference on Soil Mechanics and Foundation Engineering, Rotterdam,
vol. 5, pp 156-161.

Poulos, H.G. & Davis, E.H. (1980), Pile Foundation Analysis and Design, John Wiley & Sons.

Prakash, S. & Sharma, H.D. (1990), Pile Foundations in Engineering Practice, Wiley-IEEE.

Reese, L.C. & Van Impe, W.F. (2001), Single Piles & Groups under Lateral Loading, A.A.
Balkema, UK.

Reese, L.C. & Welch, R.C. (1975a), Lateral Loading of Deep Foundation in Stiff Clay, Journal
of Geotechnical Engineering Division, ASCE, Vol. 101 (GT7): 633-649.

Reese, L.C., Cox W.P. & Koop, F.D. (1975b), Field Testing and Analysis of Laterally Loaded
Piles in Stiff Clay. Offshore Technology Conference, Houston, Texas

Rollins, K.M., Snyder, J.L. & Broderick, R.D. (2005), “Static and Dynamic Lateral Response of
a 15 Pile Group.” Proceedings of 16th International Conference on Soil Mechanics
and Geotech. Engineering, Balkema.

Skempton, A.W. (1951), “The Bearing Capacity of Clays”, Building Research Congress,
London Institution of Civil Engineers, div.I:pp.180

Smith, T.D. & Ray, Brian (1986), “Shear Mobilization on Laterally Loaded Shafts.”
Geotechnical Aspects of Stiff and Hard Clay, New York, p. 60-68.

Smith, T.D. & Slyh, R. (1985), Lateral Friction Mobilization Rates for Laterally Loaded Piles.
38th Canadian Geotechnical Conference, Edmonton: 393-398

Tan, Y. K. and Ting, W. H. (2004). “Design of Piles Subjected to Large Lateral Soil
Movement.” Special Lecture, Proceedings of Malaysian Geotechnical Conference
2004, Kuala Lumpur, March 2004, p. 149-164.

Tonias, D.E. (1991), Bridge Engineering: Design, Rehabilitation and Maintenance of Modern
Highway Bridges, McGraw-Hill Professional.

Welch, R. C. & Reese, L. C. (1972), Laterally Loaded Behavior of Drilled Shaft, Research
Report 3-5-65-89. Center for Highway Research. University of Texas, Austin